INDEX

Explanations

locations or entities

oai_token-act-pair · gpt-3.5-turbo

special characters or symbols in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

erest

-0.77

ried

-0.69

 level

-0.66

 therap

-0.66

 lifes

-0.66

orts

-0.64

 proced

-0.62

 optimum

-0.62

 ling

-0.62

odes

-0.62

POSITIVE LOGITS

perhaps

1.04

including

1.03

 namely

1.02

meaning

0.98

––

0.97

something

0.88

they

0.87

along

0.85

particularly

0.85

even

0.84

Activations Density 0.188%

locations or entities

special characters or symbols in the text

No Comments

No Known Activations