INDEX

Explanations

the word "Sham"

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of names or terms associated with "Sham" 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 @ 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

lihood

-0.79

 Compton

-0.72

IBLE

-0.68

 Extrem

-0.65

PsyNetMessage

-0.65

 Tornado

-0.65

 Kendrick

-0.65

 Prospect

-0.64

 Quincy

-0.63

ĸļ

-0.62

POSITIVE LOGITS

eless

1.21

rock

1.17

blers

1.11

aji

1.05

Sham

0.93

ash

0.92

arie

0.92

uth

0.88

arak

0.88

robe

0.87

Activations Density 0.024%

the word "Sham"

repeated mentions of names or terms associated with "Sham" in various contexts

No Comments

No Known Activations