INDEX

Explanations

words related to specific locations or names, particularly involving fields or certain individuals

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "field" 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 @ 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

 unden

-0.81

externalActionCode

-0.80

ILA

-0.80

 undermin

-0.78

ailability

-0.78

ntil

-0.76

abase

-0.76

ccording

-0.75

 millenn

-0.75

Magikarp

-0.75

POSITIVE LOGITS

field

0.90

fields

0.83

shire

0.78

ers

0.76

 Junction

0.74

notes

0.74

er

0.73

 Marshal

0.71

hurst

0.70

VIEW

0.70

Activations Density 0.014%

words related to specific locations or names, particularly involving fields or certain individuals

occurrences of the term "field" in various contexts

No Comments

No Known Activations