INDEX

Explanations

the word "little" followed by a number indicating an extent of lack or insignificance

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "little."

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

ses

-0.66

central

-0.65

ser

-0.64

 Rear

-0.64

dor

-0.64

eneg

-0.63

RAFT

-0.63

draft

-0.63

lation

-0.63

thouse

-0.62

POSITIVE LOGITS

 consolation

0.84

 fuss

0.75

 else

0.74

rhy

0.73

 doubt

0.72

 avail

0.72

finger

0.68

 effort

0.68

bit

0.67

oooooooo

0.66

Activations Density 0.028%

the word "little" followed by a number indicating an extent of lack or insignificance

occurrences of the word "little."

No Comments

No Known Activations