INDEX

Explanations

terms related to financial and job security

oai_token-act-pair · gpt-3.5-turbo

terms related to economic security

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

lem

-0.77

INK

-0.76

phrine

-0.68

ting

-0.68

kson

-0.65

oric

-0.64

igs

-0.64

kens

-0.63

slow

-0.62

ague

-0.62

POSITIVE LOGITS

 guards

0.84

enture

0.80

Enhanced

0.78

antha

0.77

urities

0.77

riors

0.75

 guard

0.74

ailability

0.74

ilege

0.73

arial

0.71

Activations Density 0.027%

terms related to financial and job security

terms related to economic security

No Comments

No Known Activations