INDEX

Explanations

mentions of locations or directions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "along."

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

TY

-0.72

oric

-0.66

null

-0.65

meg

-0.65

BILITIES

-0.64

ptive

-0.60

Spot

-0.59

olute

-0.59

ulu

-0.59

 pitted

-0.59

POSITIVE LOGITS

side

0.81

wagon

0.80

 wagon

0.74

gradient

0.70

 Route

0.70

ward

0.65

along

0.65

agons

0.65

erous

0.64

 corridors

0.64

Activations Density 0.019%

mentions of locations or directions

occurrences of the word "along."

No Comments

No Known Activations