INDEX

Explanations

specific details or information mentioned in a text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "details" or its variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

asus

-0.79

nder

-0.74

ONY

-0.70

ammers

-0.70

azar

-0.69

rade

-0.69

sw

-0.66

arte

-0.66

ony

-0.64

onz

-0.64

POSITIVE LOGITS

 details

1.10

 detail

0.95

displayText

0.91

 glean

0.85

 redacted

0.83

 thereof

0.83

bourg

0.80

 surrounding

0.79

TBA

0.76

 Details

0.76

Activations Density 0.030%

specific details or information mentioned in a text

occurrences of the word "details" or its variants

No Comments

No Known Activations