INDEX

Explanations

words related to testing or experimentation

oai_token-act-pair · gpt-3.5-turbo

references to the act of testing or evaluating

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

taboola

-0.74

hold

-0.71

mpeg

-0.67

lihood

-0.66

thus

-0.66

joining

-0.65

heading

-0.65

zzi

-0.65

nick

-0.64

wikipedia

-0.64

POSITIVE LOGITS

osterone

1.18

imony

0.90

wcsstore

0.80

 testing

0.77

 Testing

0.75

imon

0.72

 tested

0.72

rador

0.71

 test

0.70

ipers

0.69

Activations Density 0.027%

words related to testing or experimentation

references to the act of testing or evaluating

No Comments

No Known Activations