INDEX

Explanations

verbs related to identifying people, things, or concepts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "identify" and its variations

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

orld

-0.87

ework

-0.75

perty

-0.71

bare

-0.71

fare

-0.69

 helicop

-0.68

imeo

-0.66

 bitten

-0.66

 besie

-0.64

 acquitted

-0.63

POSITIVE LOGITS

ãĤ¿

0.77

 Disorders

0.77

 è£ıè

0.75

pointers

0.73

ively

0.72

 Identification

0.72

ahime

0.71

river

0.70

ãĤ©

0.70

idable

0.67

Activations Density 0.032%

verbs related to identifying people, things, or concepts

occurrences of the word "identify" and its variations

No Comments

No Known Activations