INDEX

Explanations

phrases related to physical actions or objects

oai_token-act-pair · gpt-3.5-turbo

variations of consonant clusters or digraphs in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

LIST

-0.71

SPONSORED

-0.71

STAR

-0.69

SER

-0.69

MENT

-0.66

 Sans

-0.66

POST

-0.65

Military

-0.65

è£ħ

-0.64

PAC

-0.64

POSITIVE LOGITS

agging

0.91

ricks

0.91

apped

0.90

acker

0.89

azing

0.86

apping

0.86

acking

0.86

appings

0.86

ippers

0.85

umbing

0.85

Activations Density 0.120%

phrases related to physical actions or objects

variations of consonant clusters or digraphs in various contexts

No Comments

No Known Activations