INDEX

Explanations

adjectives and verbs indicating strong emotional impact or seriousness

oai_token-act-pair · gpt-3.5-turbo

phrases and words indicating significant themes or consequences in narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

paces

-0.87

hers

-0.78

pace

-0.75

haw

-0.72

aeus

-0.71

puting

-0.70

=~

-0.70

PUT

-0.69

hops

-0.67

wrap

-0.67

POSITIVE LOGITS

ly

1.25

 irony

1.10

 rebuke

0.93

LY

0.89

 antic

0.88

 understatement

0.88

est

0.88

 ironic

0.87

itud

0.86

ulously

0.84

Activations Density 0.417%

adjectives and verbs indicating strong emotional impact or seriousness

phrases and words indicating significant themes or consequences in narratives

No Comments

No Known Activations