INDEX

Explanations

mentions of credibility or crediting in written texts

oai_token-act-pair · gpt-3.5-turbo

references to credibility and its variations

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

 Resolution

-0.71

Ò

-0.70

hap

-0.64

AF

-0.63

manship

-0.63

 combatants

-0.62

 Franks

-0.60

 orphans

-0.59

agents

-0.58

OWER

-0.58

POSITIVE LOGITS

ulla

0.96

ey

0.94

lio

0.91

ulously

0.86

ei

0.80

ulent

0.79

lia

0.78

enza

0.77

icated

0.75

VERTISEMENT

0.74

Activations Density 0.044%

mentions of credibility or crediting in written texts

references to credibility and its variations

No Comments

No Known Activations