INDEX

Explanations

phrases starting with "Where"

oai_token-act-pair · gpt-3.5-turbo

the word "Where" used in 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

 roller

-0.73

)].

-0.68

ÂłÂłÂłÂłÂłÂłÂłÂł

-0.66

 franc

-0.60

 digest

-0.60

ATURE

-0.60

 squeeze

-0.60

 secretion

-0.60

 absorbing

-0.59

wave

-0.59

POSITIVE LOGITS

abouts

1.44

upon

1.35

fore

1.32

soever

1.07

ver

1.01

verages

0.85

nces

0.83

 else

0.78

icularly

0.76

onga

0.76

Activations Density 0.040%

phrases starting with "Where"

the word "Where" used in various contexts

No Comments

No Known Activations