INDEX

Explanations

words related to political and governmental contexts

oai_token-act-pair · gpt-3.5-turbo

references related to administrative or governmental topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 decomp

-0.97

 Codec

-0.73

gib

-0.71

hob

-0.69

 pile

-0.67

 silhou

-0.67

 sacrific

-0.67

 recording

-0.67

 parachute

-0.67

 sandwich

-0.66

POSITIVE LOGITS

£

0.94

Â¢

0.93

Ħ¢

0.90

¬

0.90

ı

0.87

ates

0.87

º

0.85

Â§

0.85

lege

0.83

ħ

0.83

Activations Density 0.376%

words related to political and governmental contexts

references related to administrative or governmental topics

No Comments

No Known Activations