INDEX

Explanations

instances of the word "string"

oai_token-act-pair · gpt-3.5-turbo

references to sequences or lists of items

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

tical

-0.89

undai

-0.69

espie

-0.67

hammad

-0.66

icago

-0.66

mos

-0.65

mares

-0.63

scl

-0.63

hemat

-0.62

psy

-0.59

POSITIVE LOGITS

ency

0.93

ently

0.88

entially

0.86

 bikini

0.85

angle

0.77

angled

0.76

encies

0.76

tie

0.73

ify

0.72

Builder

0.71

Activations Density 0.041%

instances of the word "string"

references to sequences or lists of items

No Comments

No Known Activations