INDEX

Explanations

phrases indicating surprise or strong emphasis

oai_token-act-pair · gpt-3.5-turbo

phrases indicating negation or denial

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

RAFT

-0.70

roxy

-0.63

ULTS

-0.61

ousand

-0.60

Posts

-0.60

 Comes

-0.59

inese

-0.57

rox

-0.56

perse

-0.56

CENT

-0.56

POSITIVE LOGITS

xious

1.23

 longer

1.17

ct

0.98

except

0.91

 doubt

0.91

 matter

0.84

 exception

0.84

 indication

0.83

otrop

0.82

 exaggeration

0.77

Activations Density 0.044%

phrases indicating surprise or strong emphasis

phrases indicating negation or denial

No Comments

No Known Activations