INDEX

Explanations

phrases or keywords related to physical or metaphorical limits or borders

oai_token-act-pair · gpt-3.5-turbo

references to limits or divisions in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

orah

-0.81

obe

-0.80

ocker

-0.76

imar

-0.74

atto

-0.73

otos

-0.73

ulz

-0.73

aughed

-0.71

thora

-0.70

 istg

-0.70

POSITIVE LOGITS

 boundaries

1.10

 boundary

0.93

 borders

0.87

 Limits

0.86

 Bound

0.82

 delim

0.82

 bounds

0.81

 deline

0.81

 separating

0.81

layer

0.80

Activations Density 0.037%

phrases or keywords related to physical or metaphorical limits or borders

references to limits or divisions in various contexts

No Comments

No Known Activations