INDEX

Explanations

adjectives and nouns related to abstract concepts and beliefs

oai_token-act-pair · gpt-3.5-turbo

terms and concepts related to philosophical and ideological discussions

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

KEN

-0.76

urion

-0.76

gov

-0.75

Corp

-0.75

tower

-0.73

lain

-0.72

bsite

-0.70

boarding

-0.70

etts

-0.69

ortium

-0.69

POSITIVE LOGITS

 prowess

1.09

 significance

1.06

 superiority

1.02

 realities

1.01

 impossibility

0.99

 differences

0.97

 sophistication

0.96

ities

0.95

 aspects

0.93

 implications

0.91

Activations Density 0.141%

adjectives and nouns related to abstract concepts and beliefs

terms and concepts related to philosophical and ideological discussions

No Comments

No Known Activations