INDEX

Explanations

statistics or figures within a larger context

oai_token-act-pair · gpt-3.5-turbo

references to specific entities or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Tes

-0.68

sed

-0.68

arius

-0.68

Compat

-0.66

nor

-0.60

zed

-0.59

FontSize

-0.59

ctor

-0.59

Avg

-0.58

ONSORED

-0.58

POSITIVE LOGITS

 they

0.67

 there

0.66

reau

0.62

thood

0.61

uckle

0.60

orescence

0.58

one

0.58

essional

0.57

 consisted

0.57

we

0.57

Activations Density 0.027%

statistics or figures within a larger context

references to specific entities or groups

No Comments

No Known Activations