INDEX

Explanations

updates made to articles or documents

oai_token-act-pair · gpt-3.5-turbo

occurrences of phrases that indicate updates or changes in content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hya

-0.63

 alone

-0.62

 exited

-0.61

rising

-0.61

standing

-0.58

zin

-0.58

 tongues

-0.58

 dearly

-0.57

Reviewer

-0.57

 learnt

-0.57

POSITIVE LOGITS

 accommodate

1.39

 reflect

1.35

 include

1.13

 incorporate

1.11

 coincide

1.09

 avoid

1.08

 exclude

1.07

 conform

1.06

 ensure

1.06

 maximize

1.05

Activations Density 0.141%

updates made to articles or documents

occurrences of phrases that indicate updates or changes in content

No Comments

No Known Activations