INDEX

Explanations

mentions of the word "ent"

oai_token-act-pair · gpt-3.5-turbo

words related to a specific noun or entity type

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

issance

-0.80

phrine

-0.73

BILITIES

-0.68

 Scythe

-0.65

hiba

-0.65

iets

-0.64

ichick

-0.64

ohn

-0.63

acca

-0.63

mith

-0.63

POSITIVE LOGITS

ropy

1.08

acles

1.06

ucky

1.03

ral

1.02

reprene

0.95

ertain

0.92

imental

0.91

acion

0.91

rance

0.90

inel

0.86

Activations Density 0.027%

mentions of the word "ent"

words related to a specific noun or entity type

No Comments

No Known Activations