INDEX

Explanations

specific routes or directions mentioned in text

oai_token-act-pair · gpt-3.5-turbo

references to specific routes and pathways

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

uster

-0.90

luaj

-0.78

iaries

-0.76

oat

-0.75

 Hots

-0.75

obil

-0.73

ersen

-0.72

estation

-0.72

ternity

-0.72

endi

-0.71

POSITIVE LOGITS

 routes

1.13

 route

1.06

route

1.01

Route

0.95

 Route

0.87

finding

0.87

finder

0.81

nels

0.79

 routing

0.76

 Rout

0.76

Activations Density 0.015%

specific routes or directions mentioned in text

references to specific routes and pathways

No Comments

No Known Activations