INDEX

Explanations

contexts where a significant change or increase occurs

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate significant changes or increases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

tein

-0.89

rity

-0.76

sburgh

-0.76

icip

-0.71

busters

-0.69

nan

-0.68

nar

-0.67

imir

-0.67

orno

-0.67

"}],"

-0.65

POSITIVE LOGITS

 effected

0.76

 altering

0.74

ãĥ£

0.72

 alter

0.72

 proport

0.69

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.68

 changed

0.68

owered

0.68

 impacting

0.68

 alters

0.68

Activations Density 0.015%

contexts where a significant change or increase occurs

phrases that indicate significant changes or increases

No Comments

No Known Activations