INDEX

Explanations

numerical values expressed in any digit or word form

oai_token-act-pair · gpt-3.5-turbo

references to recent events or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

é¾

-1.03

 cabbage

-0.91

 Kris

-0.84

leck

-0.82

-0.81

jit

-0.79

 Krist

-0.79

 Mayer

-0.79

-0.76

Pie

-0.76

POSITIVE LOGITS

 Extras

1.06

 unavailable

0.96

 exclus

0.95

 Underground

0.95

exempt

0.94

 indirect

0.92

 External

0.92

 inaccessible

0.90

 accessible

0.90

 unex

0.89

Activations Density 0.373%

numerical values expressed in any digit or word form

references to recent events or actions

No Comments

No Known Activations