INDEX

Explanations

words related to actions or processes

oai_token-act-pair · gpt-3.5-turbo

action-oriented words related to systemic processes and configurations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 suppose

-0.74

osuke

-0.68

 anchored

-0.64

 guid

-0.61

terday

-0.56

engu

-0.54

uden

-0.54

},"

-0.53

 Skydragon

-0.52

 Looking

-0.52

POSITIVE LOGITS

lessly

1.24

ulate

1.03

ize

0.93

uate

0.93

matically

0.91

ify

0.90

rily

0.90

istically

0.90

ishly

0.90

imize

0.84

Activations Density 0.468%

words related to actions or processes

action-oriented words related to systemic processes and configurations

No Comments

No Known Activations