INDEX

Explanations

phrases related to the Cold War

oai_token-act-pair · gpt-3.5-turbo

references to the Cold War

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

obook

-0.74

atin

-0.72

 confir

-0.69

otto

-0.66

sembly

-0.65

ewitness

-0.63

Label

-0.63

 ingred

-0.63

Asset

-0.63

GROUP

-0.61

POSITIVE LOGITS

lords

1.01

 Patton

0.92

era

0.91

lord

0.89

Era

0.87

riors

0.82

 hysteria

0.77

naire

0.76

 waged

0.75

 raging

0.75

Activations Density 0.031%

phrases related to the Cold War

references to the Cold War

No Comments

No Known Activations