INDEX

Explanations

verbs indicating belief or speculation

oai_token-act-pair · gpt-3.5-turbo

expressions of suspicion or conjecture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

perty

-0.84

ciating

-0.73

arantine

-0.72

iphate

-0.71

ebra

-0.69

Synopsis

-0.64

elight

-0.63

thinkable

-0.63

osc

-0.63

eatured

-0.62

POSITIVE LOGITS

 whoever

0.85

 Redmond

0.79

 readers

0.76

 that

0.76

 most

0.75

 Rasmussen

0.71

CCP

0.70

it

0.70

 many

0.69

 this

0.66

Activations Density 0.128%

verbs indicating belief or speculation

expressions of suspicion or conjecture

No Comments

No Known Activations