INDEX

Explanations

words related to explosive or destructive events

oai_token-act-pair · gpt-3.5-turbo

terms related to various types of acid or chemical substances

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

ĸļ

-0.85

nah

-0.78

different

-0.72

 warr

-0.69

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.69

sold

-0.69

witz

-0.69

 Taken

-0.68

 Varg

-0.67

pherd

-0.67

POSITIVE LOGITS

ALLY

1.00

otine

0.97

inity

0.90

entric

0.89

henko

0.88

ulum

0.87

 acid

0.85

ulation

0.84

ulous

0.83

orp

0.82

Activations Density 0.042%

words related to explosive or destructive events

terms related to various types of acid or chemical substances

No Comments

No Known Activations