INDEX

Explanations

phrases related to documented facts or events

oai_token-act-pair · gpt-3.5-turbo

instances of the word "documented" and its variations, indicating discussions about evidence or record-keeping

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

awar

-0.87

enta

-0.77

sv

-0.76

adding

-0.75

mun

-0.74

yss

-0.73

spr

-0.72

cel

-0.72

isoft

-0.71

egal

-0.69

POSITIVE LOGITS

 documenting

0.92

 documented

0.91

documented

0.89

ocument

0.82

 documentation

0.78

ĸļ

0.73

urally

0.73

iary

0.69

 feats

0.69

URES

0.68

Activations Density 0.010%

phrases related to documented facts or events

instances of the word "documented" and its variations, indicating discussions about evidence or record-keeping

No Comments

No Known Activations