INDEX

Explanations

locations or points of entry

oai_token-act-pair · gpt-3.5-turbo

references to entrances and exits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rior

-0.74

nian

-0.69

olin

-0.66

irit

-0.66

amera

-0.65

conn

-0.65

ilot

-0.64

akia

-0.61

 inherit

-0.61

verbs

-0.61

POSITIVE LOGITS

 thereto

0.97

prise

0.96

tainment

0.90

 portals

0.90

ways

0.87

prises

0.86

 entrances

0.86

 gates

0.85

 entrance

0.85

 portal

0.80

Activations Density 0.031%

locations or points of entry

references to entrances and exits

No Comments

No Known Activations