INDEX

Explanations

dates in a specific format, possibly related to events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "initial" 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

Sav

-0.77

knife

-0.72

yers

-0.71

rf

-0.70

romeda

-0.70

Fish

-0.69

rod

-0.68

bara

-0.67

lov

-0.67

razil

-0.67

POSITIVE LOGITS

 stages

1.06

 impressions

1.02

 impression

0.97

 impulse

0.92

 reaction

0.91

 phases

0.90

 responders

0.90

 batch

0.90

izers

0.89

 impetus

0.88

Activations Density 0.025%

dates in a specific format, possibly related to events

occurrences of the word "initial" in various contexts

No Comments

No Known Activations