INDEX

Explanations

words related to experiences and activities

oai_token-act-pair · gpt-3.5-turbo

expressions of emotional experiences and personal reflections

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

vernment

-0.66

 POLITICO

-0.65

'?

-0.63

"?

-0.61

morrow

-0.59

exit

-0.59

 Conservatives

-0.58

 Wednesday

-0.58

 Judicial

-0.57

gencies

-0.57

POSITIVE LOGITS

okingly

0.64

 attest

0.61

nown

0.60

 unfortunately

0.60

entimes

0.60

found

0.60

 fused

0.58

 Vaugh

0.58

 synergy

0.58

Introduced

0.58

Activations Density 1.374%

words related to experiences and activities

expressions of emotional experiences and personal reflections

No Comments

No Known Activations