INDEX

Explanations

mentions of collaborative decision-making and discussions as well as verification and feedback

oai_token-act-pair · gpt-3.5-turbo

statements related to feedback and approval in conversations or discussions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤ¼ãĤ¦ãĤ¹

-0.66

illac

-0.63

activation

-0.61

 Activate

-0.61

inational

-0.61

Located

-0.60

 Battalion

-0.60

stasy

-0.60

ufact

-0.60

veyard

-0.59

POSITIVE LOGITS

 critiques

1.09

 summarize

1.04

 conclusions

1.03

 critique

1.02

 opinions

1.00

 readers

1.00

 enlight

1.00

 replies

0.99

 summar

0.98

 comments

0.98

Activations Density 1.060%

mentions of collaborative decision-making and discussions as well as verification and feedback

statements related to feedback and approval in conversations or discussions

No Comments

No Known Activations