INDEX

Explanations

the acronym "EE" at various positions within phrases or sentences

oai_token-act-pair · gpt-3.5-turbo

references to electrical engineering or associated institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

otine

-0.93

ifiers

-0.78

ique

-0.77

ifier

-0.75

icum

-0.74

ication

-0.74

iewicz

-0.73

nikov

-0.72

ifiable

-0.69

eus

-0.68

POSITIVE LOGITS

TY

1.05

EEE

0.97

KS

0.92

KN

0.91

LECT

0.91

VO

0.89

VEL

0.84

FAULT

0.83

ENCY

0.83

FI

0.82

Activations Density 0.009%

the acronym "EE" at various positions within phrases or sentences

references to electrical engineering or associated institutions

No Comments

No Known Activations