INDEX

Explanations

words related to achievements or distinctions

oai_token-act-pair · gpt-3.5-turbo

descriptive terms indicating significance or ranking

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Okin

-0.74

 Allaah

-0.72

 Yourself

-0.72

 Poles

-0.69

rarily

-0.69

lov

-0.66

 Bett

-0.64

ones

-0.64

 Topic

-0.63

KING

-0.62

POSITIVE LOGITS

 foray

0.93

ãĥķãĤ©

0.82

 selves

0.82

 counterparts

0.82

 flagship

0.81

 holdings

0.80

 counterpart

0.80

 estate

0.77

 predicament

0.75

 inaugural

0.74

Activations Density 0.217%

words related to achievements or distinctions

descriptive terms indicating significance or ranking

No Comments

No Known Activations