INDEX

Explanations

the presence of the name "Evans" in various contexts

oai_token-act-pair · gpt-3.5-turbo

the name "Evans" as it appears frequently in relation to 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

MY

-0.76

PM

-0.75

 spare

-0.72

roxy

-0.70

gio

-0.68

scribed

-0.68

yright

-0.68

interstitial

-0.66

ariat

-0.66

pees

-0.65

POSITIVE LOGITS

olution

0.87

olve

0.82

ville

0.80

rolet

0.78

burgh

0.74

angel

0.74

 Evans

0.74

 Knox

0.73

aunders

0.71

olver

0.71

Activations Density 0.007%

the presence of the name "Evans" in various contexts

the name "Evans" as it appears frequently in relation to various contexts

No Comments

No Known Activations