INDEX

Explanations

words ending with 'nex' or 'ause'

oai_token-act-pair · gpt-3.5-turbo

words associated with connections and links, particularly in contexts involving networks or relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

TEXTURE

-0.90

rawdownloadcloneembedreportprint

-0.73

SET

-0.69

 Mata

-0.67

 Lump

-0.66

asso

-0.65

 Shepard

-0.64

 Nanto

-0.64

 Offer

-0.63

EStreamFrame

-0.62

POSITIVE LOGITS

ause

0.98

uve

0.82

velop

0.80

riad

0.77

nex

0.77

oard

0.76

ive

0.74

opausal

0.73

ird

0.72

ighty

0.72

Activations Density 0.027%

words ending with 'nex' or 'ause'

words associated with connections and links, particularly in contexts involving networks or relationships

No Comments

No Known Activations