INDEX

Explanations

phrases related to comparison or evaluation

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize negation or denial

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Almighty

-0.64

LIN

-0.63

 Naples

-0.59

WT

-0.59

Condition

-0.56

catentry

-0.56

multipl

-0.56

 motif

-0.55

TEXTURE

-0.54

soType

-0.53

POSITIVE LOGITS

 anymore

0.69

vae

0.68

nor

0.64

 Enough

0.64

 bother

0.63

dfx

0.61

REDACTED

0.61

 Cheong

0.61

userc

0.60

cffff

0.59

Activations Density 0.182%

phrases related to comparison or evaluation

phrases that emphasize negation or denial

No Comments

No Known Activations