INDEX

Explanations

denials or refutations of accusations or allegations

oai_token-act-pair · gpt-3.5-turbo

statements of denial or rejection

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

=-=-=-=-=-=-=-=-

-0.89

incinn

-0.85

================================

-0.77

=-=-=-=-

-0.76

OPER

-0.75

Compar

-0.75

MAT

-0.73

ARCH

-0.71

Center

-0.70

Nev

-0.69

POSITIVE LOGITS

 denies

1.00

 denying

1.00

 deny

0.94

 denied

0.93

 denial

0.93

 outright

0.81

 excuses

0.73

 dismissing

0.70

 McMaster

0.69

zzle

0.67

Activations Density 0.016%

denials or refutations of accusations or allegations

statements of denial or rejection

No Comments

No Known Activations