INDEX

Explanations

mentions of the word "History"

oai_token-act-pair · gpt-3.5-turbo

references to "history" and its various aspects or contexts

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

aton

-0.75

inki

-0.74

leased

-0.74

ients

-0.73

 medically

-0.72

ressed

-0.71

ached

-0.70

orc

-0.69

nery

-0.69

ividual

-0.68

POSITIVE LOGITS

 buffs

0.83

ISTORY

0.80

 History

0.78

making

0.77

 Lith

0.76

 history

0.75

 nostalg

0.75

 tale

0.73

 textbooks

0.70

 Fiction

0.70

Activations Density 0.027%

mentions of the word "History"

references to "history" and its various aspects or contexts

No Comments

No Known Activations