INDEX

Explanations

mentions of the word "Reading" which can refer to a location or an act of reading

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Reading."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Â¯Â¯Â¯Â¯

-0.94

arb

-0.84

ĪĴ

-0.81

lando

-0.81

obin

-0.79

cial

-0.75

hl

-0.75

ofi

-0.75

selage

-0.75

gger

-0.75

POSITIVE LOGITS

 Reading

0.80

sburg

0.79

 Rainbow

0.77

 Through

0.75

ford

0.74

 aloud

0.73

 Passage

0.73

 Eleven

0.71

 comprehension

0.71

 Journals

0.71

Activations Density 0.005%

mentions of the word "Reading" which can refer to a location or an act of reading

instances of the word "Reading."

No Comments

No Known Activations

mentions of the word "Reading" which can refer to a location or an act of reading

instances of the word "Reading."

No Comments

No Known Activations