INDEX

Explanations

phrases related to problem-solving or decision-making

oai_token-act-pair · gpt-3.5-turbo

phrases related to figuring things out

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

avorite

-0.79

cious

-0.77

cius

-0.76

interstitial

-0.75

asus

-0.65

onds

-0.65

agos

-0.65

wake

-0.64

esa

-0.64

 repro

-0.64

POSITIVE LOGITS

wards

0.85

posts

0.84

fitted

0.81

how

0.76

OTAL

0.76

casts

0.75

lier

0.74

tical

0.70

omorphic

0.68

\\\\\\\\

0.67

Activations Density 0.046%

phrases related to problem-solving or decision-making

phrases related to figuring things out

No Comments

No Known Activations