INDEX

Explanations

summaries or highlights of information

oai_token-act-pair · gpt-3.5-turbo

references to summaries across 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

train

-0.73

hips

-0.72

 charism

-0.72

yss

-0.71

bees

-0.71

gypt

-0.70

warm

-0.66

apy

-0.65

bat

-0.64

Wee

-0.63

POSITIVE LOGITS

 summ

0.85

 summary

0.83

 summar

0.80

 synopsis

0.79

 summarize

0.77

krit

0.76

strate

0.75

 VIEW

0.75

 outline

0.72

ãĤ¢ãĥ«

0.72

Activations Density 0.023%

summaries or highlights of information

references to summaries across various contexts

No Comments

No Known Activations