INDEX

Explanations

words related to influence and impact

oai_token-act-pair · gpt-3.5-turbo

expressions of connection and relational dynamics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 usually

-0.68

 typically

-0.63

igi

-0.62

 averaging

-0.61

Reviewer

-0.61

idav

-0.58

 hopefully

-0.58

assert

-0.57

 gotta

-0.57

igers

-0.57

POSITIVE LOGITS

omorphic

0.80

 Kislyak

0.77

untled

0.66

 PsyNet

0.62

ãĤ¨ãĥ«

0.62

 conflic

0.61

ammed

0.60

 extrater

0.59

anism

0.59

ãĥĥãĤ¯

0.58

Activations Density 0.487%

words related to influence and impact

expressions of connection and relational dynamics

No Comments

No Known Activations