INDEX

Explanations

adjectives or nouns representing characteristics or attributes

oai_token-act-pair · gpt-3.5-turbo

adjectives and descriptive terms

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

owship

-1.03

olicited

-0.93

tower

-0.78

IRD

-0.77

inki

-0.75

rir

-0.74

ãģ®éŃĶ

-0.73

mouth

-0.72

under

-0.72

ortium

-0.72

POSITIVE LOGITS

ative

0.85

 advers

0.70

 easing

0.69

 mating

0.69

 measures

0.69

Pwr

0.69

 minded

0.67

 matter

0.67

 effect

0.66

 Powers

0.65

Activations Density 0.046%

adjectives or nouns representing characteristics or attributes

adjectives and descriptive terms

No Comments

No Known Activations