INDEX

Explanations

the word "Yet" at the beginning of sentences

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Yet."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

tein

-0.85

tained

-0.77

atti

-0.74

packs

-0.71

machine

-0.69

rities

-0.69

omial

-0.69

broken

-0.68

rador

-0.67

strings

-0.67

POSITIVE LOGITS

tons

0.87

 somehow

0.83

 alas

0.80

theless

0.78

 strangely

0.76

 despite

0.72

heric

0.70

entimes

0.69

 mirac

0.69

 inexpl

0.66

Activations Density 0.014%

the word "Yet" at the beginning of sentences

instances of the word "Yet."

No Comments

No Known Activations