INDEX

Explanations

fragments or incomplete pieces

oai_token-act-pair · gpt-3.5-turbo

words related to fragments or fragmentation

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

reon

-0.78

know

-0.77

OPA

-0.74

Keefe

-0.72

rol

-0.71

kins

-0.70

orney

-0.69

perature

-0.68

Demand

-0.68

sburgh

-0.68

POSITIVE LOGITS

 fragments

1.12

 fragment

0.93

 Frag

0.89

 ILCS

0.85

atis

0.84

 fragmented

0.81

mentation

0.80

bite

0.79

glass

0.79

 shards

0.76

Activations Density 0.020%

fragments or incomplete pieces

words related to fragments or fragmentation

No Comments

No Known Activations