INDEX

Explanations

terms related to peak values or operations

oai_token-act-pair · gpt-3.5-turbo

numerical data associated with thresholds, limits, or significant events

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

phia

-0.75

enhagen

-0.73

ibly

-0.68

riend

-0.65

kind

-0.63

itte

-0.62

theless

-0.62

cient

-0.60

pell

-0.59

cience

-0.59

POSITIVE LOGITS

 Ferry

0.59

hedon

0.57

scl

0.56

 Destination

0.56

 Alas

0.55

 Landing

0.55

 Nose

0.53

 boulder

0.53

endment

0.53

 Sail

0.53

Activations Density 0.993%

terms related to peak values or operations

numerical data associated with thresholds, limits, or significant events

No Comments

No Known Activations