INDEX

Explanations

words related to physical illness or discomfort

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "sick" in various contexts

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

 unlaw

-0.69

 sanctioned

-0.67

 Unch

-0.66

 principals

-0.65

 compr

-0.64

 Goodwin

-0.62

wcsstore

-0.61

 guid

-0.61

CLS

-0.60

rul

-0.59

POSITIVE LOGITS

ening

1.47

ened

1.31

bay

1.27

er

0.98

nesses

0.96

estro

0.93

ert

0.92

erton

0.92

ly

0.90

ness

0.90

Activations Density 0.024%

words related to physical illness or discomfort

occurrences of the word "sick" in various contexts

No Comments

No Known Activations